Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Characterization of mammographic masses using a gradient-based segmentation algorithm and a neural classifier.

Pasquale Delogu1, Maria Evelina Fantacci, Parnian Kasae

  • 1Dipartimento di Fisica dell'Università di Pisa and INFN Sezionedi Pisa, Largo Pontecorvo 3, 56127 Pisa, Italy.

Computers in Biology and Medicine
|March 27, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A digital twin approach for simultaneous reconstruction of brain anatomy and dynamics from neural data.

PLOS digital health·2026
Same author

A coincidence-based response matrix for correction of charge sharing spectral distortions in photon counting detectors.

Medical physics·2026
Same author

Denoising of low-dose chest computed tomography images using a U-net based convolutional autoencoder and transfer learning.

Biomedical physics & engineering express·2026
Same author

Improving patient treatment accuracy using transit dosimetry with Electronic Portal Imaging Device images and deep learning.

Physics and imaging in radiation oncology·2026
Same author

Efficient machine learning models leveraging DCE-MRI morphological and dynamic features allow accurate breast lesion classification.

Biomedical physics & engineering express·2026
Same author

Corrigendum: Phase-contrast breast CT: the effect of propagation distance (2018<i>Phys. Med. Biol</i>.<b>63</b>24NT03).

Physics in medicine and biology·2026

A novel computer-aided diagnosis (CAD) system effectively characterizes mammographic masses using a gradient-based segmentation algorithm and neural network classification. This automated system shows strong potential for improving diagnostic accuracy in breast cancer screening.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Mammography is crucial for breast cancer screening.
  • Accurate characterization of mammographic masses is vital for diagnosis.
  • Computer-aided diagnosis (CAD) systems can assist radiologists.

Purpose of the Study:

  • To develop and evaluate a novel CAD system for mammographic mass characterization.
  • To assess the system's performance in distinguishing malignant from benign masses.
  • To investigate the impact of feature selection on classification accuracy.

Main Methods:

  • A gradient-based segmentation algorithm was developed for mass identification.
  • Sixteen features (shape, size, intensity) were extracted from segmented masses.

Related Experiment Videos

  • A multi-layered perceptron neural network, trained via back-propagation, was used for classification.
  • Receiver-Operating Characteristic (ROC) analysis was employed to evaluate performance.

    • Main Results:

    • The segmentation algorithm demonstrated comparable efficiency on both malignant and benign masses.
    • A feature selection process identified 12 optimal features out of 16 for best classifier performance.
    • The CAD system achieved an area under the ROC curve (A(z)) of 0.805+/-0.030 for correctly segmented masses.
    • Overall system performance, including acceptably segmented masses, resulted in A(z)=0.780+/-0.023.

    Conclusions:

    • The developed CAD system shows significant potential for assisting radiologists in mammographic mass diagnosis.
    • The gradient-based segmentation and neural network classification approach is robust and adaptable.
    • Optimized feature selection enhances the diagnostic accuracy of the CAD system.